0

I am trying to implement a Software in the loop simulation, I do not find the issue here. I am currently debugging using constant input vectors. Appears the problem is given by the allocation and definition of u1, the output variable.

What am I doing wrong?

The code works in a Function MATLAB evnironment, but the same does not work in simulink MATLAB Function Block.

function [u1,deltau]  = Controller(Omega,Psi,Lzerot,xm,y,sp,m)

persistent Xf
if isempty(Xf)
Xf = [xm;(y-sp)];    
end

persistent xm_old
if isempty(xm_old)
xm_old = xm;
end

persistent u
if isempty(u)
u = zeros(m,1);
end

persistent deltau_old
if isempty(deltau_old)
deltau_old = zeros(m,1);
end

% O P T I M I Z A T I O N
eta           = -(Omega\Psi)*(Xf);  % (L. Wang pg. 99/100)
deltau        = Lzerot*eta;         % Kmpc = L(0)'*(Omega\Psi) & Du = - Kmpc * x(k)

if deltau == deltau_old
else
    u = u + deltau;
    deltau_old = deltau;

end

u1 = u;

% U P D A T E 
if xm == xm_old
% elseif y == y_old
%     Xf      = [xm-xm_old;(y-sp)];
else 
    Xf      = [xm-xm_old;(y-sp)];
    xm_old  = xm;
end


end

Thank you for your help.

I tried to pre-allocate the u1 variable, as (m,1) vector but it does not affect the problem. Cleaning the function and optimizing is part of the task I want to achieve. Is it just that I need the zero allocation of u, for that reason I am using the persistant type of variable.

I do not want my variable to change size. It has not to.

Even if I change to "variable size" type of output, it does not solve the problem.

ovolavola
  • 1

0 Answers0